China’s Qinglang Campaign Targets AI Chaos in Phase 2
China’s Central Cyberspace Administration Commission (CAC) has launched the second phase of its “Qinglang” (Clear and Bright) special action targeting AI application chaos, focusing on AI-generated false information, violent and vulgar content, impersonation, and content that infringes on minors’ rights, according to Xinhua News. The announcement, first published via the official “China Cyberspace” WeChat account on September 2, directs local cyberspace authorities to intensify enforcement and urges platforms to systematically clean up AI-generated illegal content.
Context: An Expanding Regulatory Framework
The second phase builds on the broader 2026 Qinglang campaign, which was launched on April 30 as a four-month nationwide effort divided into two stages. The first phase, which ran through July, targeted service-level violations including failure to file large language models, inadequate AI platform security, AI data poisoning, and non-compliance with synthetic content labeling requirements. According to the National Rumor Refutation Platform’s report on phase one results, authorities disposed of more than 14,000 illegal AI products, cleaned up over 6 million pieces of illegal information, and handled approximately 26,000 accounts during the first phase, which concluded in early July.
The Qinglang campaign series itself has been an annual internet governance exercise conducted by the CAC since 2016, becoming annual since 2020, as documented on Wikipedia. The 2026 edition represents a significant escalation from its 2025 predecessor, operating within a more developed regulatory environment that includes mandatory AIGC labeling standards that took effect in September 2025 and new interim measures for anthropomorphic AI interactive services published in April 2026.
Cumulative Enforcement Results
Across both phases of the 2026 campaign, authorities report having cleaned up more than 5.61 million pieces of illegal or irregular information, investigated and handled over 49,000 accounts, and disposed of more than 2,400 illegal websites and applications, as CCTV News and China News Service reported. The CAC’s reporting center also opened a dedicated “AI Application Chaos Reporting Zone” in June 2026 to facilitate public reporting.
Major platforms including Douyin, Kuaishou, Weibo, Tencent, Baidu, Bilibili, Xiaohongshu, Zhihu, Douban, and Taobao have collectively published 46 governance announcements. AI model providers such as Doubao, Yuanbao, Qianwen, and Wenxin Yiyan have strengthened original corpus content review and AI-generated content labeling, while app stores operated by Huawei, Xiaomi, OPPO, and vivo have established stricter developer access standards.
Phase 2 Focus: Content-Level Misuse
The second phase targets seven categories of typical cases, as detailed in the CAC announcement. These include the AI “magic modification” of classic literary works to generate low-quality sensational content, AI-generated false and unverified information including fake disaster scenes and fabricated social topics, AI impersonation of public figures through face-swapping and voice-changing technologies, mass production of violent and vulgar content, AI content that infringes on minors’ rights including so-called “poison animations,” the use of AI “managed hosting” software for online water army activities, and violations by AI products and applications themselves.
Analysis: A Coordinated Governance Approach
The two-phase structure demonstrates a comprehensive regulatory strategy that addresses both the supply side—AI developers and service providers—and the demand side—content distribution platforms, as The Next Web noted in its analysis of the 2026 campaign. The campaign operates against a backdrop of heightened international scrutiny of China’s AI ecosystem, though it remains fundamentally a domestic governance exercise focused on protecting Chinese users.
Local cyberspace authorities in Beijing, Shanghai, Zhejiang, and Guangdong have implemented region-specific measures, from upgraded content review strategies to compliance consultations with key platforms. The emphasis on protecting minors from AI-generated harmful content reflects broader Chinese policy priorities around youth protection in the digital space.
What’s Next
Cyberspace authorities have indicated they will continue to monitor new risks emerging from AI-generated synthetic content and take timely measures to prevent the abuse and malicious use of AI technology. The CAC has stated its intention to create a sustainable environment for AI innovation while maintaining regulatory oversight. As the campaign’s enforcement mechanisms mature and the regulatory framework continues to evolve, observers will be watching how China balances its stated commitment to AI innovation with increasingly assertive content governance.